te-network-monitoring

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches MCP server components from the CiscoDevNet GitHub organization and communicates with official Cisco API endpoints at api.thousandeyes.com.
  • [REMOTE_CODE_EXECUTION]: Employs npx mcp-remote for connecting to a remote MCP server hosted by Cisco.
  • [COMMAND_EXECUTION]: Includes setup instructions for environment management using git clone and pip install.
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface: (1) Ingestion points: Network test results, events, and dashboard widget data retrieved from the ThousandEyes API. (2) Boundary markers: No markers or delimiters are specified to distinguish monitored data from agent instructions. (3) Capability inventory: Tools for account management (users, groups) and active troubleshooting (instant tests). (4) Sanitization: No sanitization of the external monitoring content is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 12:32 AM